To my students, during finals week,
I regret to inform you that I cannot
- Tell you within an hour or two of your handing me your final exam what your grade for the course will be (as I need to grade about 130 of these exams and you will note that not all of the items were multiple choice),
- Grade your answers on the basis of what you meant rather than what you wrote (especially on the items that were multiple choice),
- Tell you, as you’re handing me your final exam, whether the exam grades will be curved, or if so, what the curve will look like (see above about the number of exams to be graded — and then entered into a spreadsheet to run the stats),
- Reassure you that your participation will contribute positively to your final grade if you have attended class meetings so seldom that your face rings no bells for me at all,
- Create an extra-credit assignment just for you to counteract the negative effect of your having blown off all the graded assignments besides the midterm and final exams (since it would be unfair to do so without offering your classmates the same opportunity, and since I’m already working on the very edge of what is possible just to grade the non-extra coursework to submit grades on time).
This whole exam thing, and the larger grading thing, is supposed to be about evaluating what you have learned during the semester.
I have done my best to make the material accessible and maybe even interesting. I fully recognize that it isn’t everyone’s cup of tea. And, having been a student myself for a lot of years, I understand that life sometimes unfolds in ways that make it hard to focus on school, to come to class meetings, and to get the assigned work turned in.
Among other things, college can teach us that we’re not always going to get straight A’s (or, necessarily, passing grades) during semesters we can’t get it together to do the school work.
If you earn a low grade, that doesn’t make you a bad person.
And, if I give you the grade you’ve earned, that just means I’m doing my job. I don’t always enjoy it, but that’s how it goes.
Here, here.
A quibble. We professors don’t give grades, we only record grades. If your students understand that, your life will be (one hopes) a little less hectic.
Very few students seem to grasp the concept of getting the grade you earn.
You don’t know the method by which you will be grading the papers before you start the course?
Pascale- when professors understand that students EARN their grades by PROVING their knowledge to the professors instead of by WORKING HARD or by LEARNING, then students will understand they are GETTING the grades they EARN.
becca’s on it.
Because I cannot jam electrodes into my students’ heads to get a direct measure of what they have learned (and don’t want to, and would require a truckload of information about how to connect learning the material with particular measurable levels of stuff in their brains), we have to rely on crude instruments — like exams — to try to get the relevant information.
Some exams (and exam items) are better at distinguishing real comprehension, ability to extend concepts to new situations, etc., than others.
Some students who are actually pretty good at learning the material are also really good at psyching themselves out for exams, or haven’t thought much about effective strategies for taking exams, etc.
Some students who don’t spend much time learning the course material are actually pretty sophisticated in their test-taking strategies.
It would be ideal if I could reliably construct exams that let students show me what they know without providing cover for flim-flammers. Sometimes I get closer to the ideal, other times not so much.
This is part of why I sometimes have to curve exam grades — sometimes I discover in the process of grading ’em that one section just didn’t work the way I thought it would. (I tend to have more than one kind of item testing comprehension of a particular issue on any given exam — internal controls, kind of.) In those cases, I don’t want to punish the students for a flaw in the instrument I designed with which to evaluate them.
(Other times, of course, a curve can be about mercy.)
A whole heap of likes and +1s and whatever else to the item about creating additional extra credit. My syllabus says “no extra credit whatsoever” for a reason.
Things I could do but won’t, because this should be the student’s responsibility:
* Give you a personalized re-lecture of every class you missed (there were 100 other students there that day, surely one of them will share their notes)
* Base your grade on the GPA you need to get into another major
* Give your coach a progress report on your performance in class
* Calculate what grade you need to get on the final exam to get a certain final grade in the class (you had to take basic algebra to get into college, so why not use it?)
I have never been marked “on a curve” and I have never been offered extra credit. Ever.
There is a point of view that if results of answers on a particular question do not correlate with general success on the test, that question should be thrown out. Students who got the question right do not think this is fair, of course. It is interesting to think about how to fiddle around with this.
A story based on Stentor’s last comment. When I took introductory biology the grade was 1/3 lab exams, 1/3 lecture exams, 1/3 final exam, no curve. Another fellow and I were battling it out for high A. I went absolutely blank on the final. The other fellow got the only A, and I got the high B out of 300 students.
During the second semester I did not do as well. A 100 on the final would not have gotten me an A. I did not finish the final. I turned it in to my lab TA and left. He came after me. I explained to him that I could not get and A, that 76 points would get me a B, and I had 83 points on what I had finished.
During the summer, I saw the TA. He called me over, and told me, in confidence, the grades were so bad that for the first time in history the grades were curved. I did make 83 on the final. If I had made 85 I would have received an A.
Moral. Don’t out-figure your self.
Alternate moral: Too much is always better than not enough (at least when it comes to points earned on exams).
In related news, I’ve encountered a bunch of test papers on which no answer was indicated for items in the “true/false” section. Students, I beg of you: GUESS if you don’t know the answer for a true/false item! Please trust that I have additional items (including the free-response ones that take such a long time for me to grade) to discern whether you really understood the issues about which the true/false items ask.
Moral question, based on a real-life event*:
If a student who had mostly gotten A grades during the course of the year approached you after the final exam and pointed out that they missed the final, worth 50% of the grade, because the you (the lecturer) elected to say out loud the date in those final few minutes when everyone was noisily heading out of the classroom, rather than write it on the blackboard, and they (the student), with a hearing disability, incorrectly “heard”** you – would you allow their grade to be taken from internal assessment?
One catch here, that not all realise, is that you can’t (always) “know” that you’re misheard something – you simply get it wrong entirely innocently.
I realise a cynic would point to that this opens up the possibility of the student lying; that’s why I’ve included the grades they got during the year.
* Not in Janet’s class or university, I hasten to add!
** In inverted commas, as in this situation the student will almost certainly be lip-reading only.
Is there no way to check the date of the exam? If I could not trust my hearing, I would be looking on the web for the actual date.
Perhaps you missed –
“you can’t (always) “know” that you’re misheard something – you simply get it wrong entirely innocently.”
You don’t always get a sense of ‘not trusting’ what you heard. You can simply ‘hear’ it incorrectly thinking you got it right.
I take it you have missed Janet’s later comment and my reply to it? (See below, esp. my remark that the setting was pre-WWW, etc.)
Excellent, Dr. Stemwedel.
Sincerely,
Superannuated Teacher
(btw, where on earth did the person who has “never been marked on a curve” attend school? And yes, a test is an evaluative tool; and, it is a learning tool for the teacher. So of course it is not possible to know how the “curve” may or may not be utilized until the exam is examined, so to speak.)
I attended school in the UK.
Any exam I ever took was given a %age grade.
Then I trust that you never took GCSEs or A-Levels; both of those have a weird grading system where your “percentage” mark is *not* the percentage you got right, but a percentage taken against the grading curve for that particular exam. As a result, the correlation between letter grades and percentages stays constant.
This gives you some oddities – 1 mark out of 100 possible might be 10%, while 90 marks out of 100 possible might be a mere 60%. This is done so that (in theory) the same percentage each year is the same ability each year, even as the exam changes in difficulty, and thus the grade letter you achieve is the same as anyone else who ever got the same percentage as you did.
To be fair, it is possible to be graded on a curve without knowing that this is happening (e.g., in the case where the partial credit scheme is worked out in such a way that it reflects the success of the test as a tool of evaluation and generates scores that fall into the “right” numerical bins for what would appear to be non-curved grading).
Grant, I’m thinking about the ethical issues in your example (because it’s more fun than hashing out the partial credit scheme I need to hash out for the exam item I’m grading right now). But, I’m happy to say that the situation you describe is practically impossible at my university. The registrar schedules the final exams (based on the meeting days/times of the courses) and publishes these online before the term starts. For safety’s sake, I reproduce this information in my course syllabi and websites — which means you have to make a real effort *not* to know when the exam will be. (Not that I doubt some are willing to make that effort …)
You’ve caught a point there – place my example in a setting from about 30 years ago. No WWW, and the dates/times of the final exams not set until sometime within the university year. (That’s the actual setting, FWIW.)
Educational research has found that at least 80% of TF questions should be true, or the test-taking student will begin to have doubts about their knowledge, which will lead them to make induced mistakes. I like this because I find it very difficult to write a question with a false answer.
I have suggested to students when taking a multiple guess test and being unsure of an answer, to mark B. Students have later thanked me for this advice.
On the other hand, I will never forget the cry of the student who flung open his dorm window on a Friday afternoon and howled pathetically out over the quad,
NOBODY MAKES THEM ALL TRUE
ON A TRUE AND FALSE TEST!!!!!
Dear Janet, would you mind explaining to the non-native English speaker what a “curve” is? “To be graded on a curve” .. it certainly does not mean that you receive your grade while skiing downhills, does it.
“Grading on a curve” is deciding what grades the points earned translate to by taking account of the performance of the test-takers as a group. Sometimes people do this on the basis of a strict statistical formula that is determined in advance (e.g., the mean is a C, one standard deviation up from that is a B, two standard deviations up from the mean is an A, one standard deviation down from the mean is a D, etc.). Others figure out the appropriate translation of points to grades post hoc, drawing on other information they have about how well students have understood the material, how successful a tool for evaluation the exam has turned out to be, and so forth.